Iridium-catalyzed formation of trans-polyphenylacetylene by alkyne polymerization

Article abstract of DOI:10.1016/S1381-1169(03)00415-1

The iridium(I) compounds HIr(cod)(PR₃)₂ (cod: 1,5-cyclooctadiene; PR₃: Ph₃, P(p-MeOC₆H ₄)₃, P(o-MeOC₆H₄)Ph₂, PCyPh₂, PCy₂Ph

Full text of DOI:10.1016/S1381-1169(03)00415-1

Journal of Molecular Catalysis A: Chemical 206 (2003) 319–329
Iridium-catalyzed formation of trans-polyphenylacetylene
by alkyne polymerization
Mauro Marigo, Dunja Millos, Nazario Marsich, Erica Farnetti∗
Dipartimento di Scienze Chimiche, Universita’ di Trieste, Via L. Giorgieri 1, I-34127 Trieste, Italy
Received 20 February 2003; accepted 14 May 2003
Abstract
The iridium(I) compounds HIr(cod)(PR3)2 (cod: 1,5-cyclooctadiene; PR3: Ph3, P(p-MeOC6H4)3, P(o-MeOC6H4)Ph2,
PCyPh2, PCy2Ph) were employed as catalyst precursors for the polymerization of phenylacetylene. The polyene was formed
as the major product with all the catalysts except the PCy2Ph derivative, which promoted preferential formation of oligomers.
In all cases the polymerization reactions were highly stereoselective, yielding 100% trans-polyphenylacetylene. From the
≡
catalytic mixtures the iridium(III) derivatives fac-HIr(C CPh)2(PR3)3 (PR3: PPh3, P(p-MeOC6H4)3) were isolated. The
results of spectroscopic studies are also reported, which provide information on the evolution of the iridium precursors during
the catalytic reaction.
© 2003 Elsevier B.V. All rights reserved.
Keywords: Iridium; Phosphines; Homogeneous catalysis; Alkynes; Polymerization
1. Introduction
num and tungsten-based catalysts promote the forma-
tion of polyenes with high molecular weight, in some
cases in a living manner [5–9]. More recently, nickel
and palladium-based catalysts have been reported to
be active in alkynes polymerization [10,11]. On the
other hand, organorhodium derivatives form another
class of catalytic systems which have been studied
by several groups [12–20], obtaining excellent results
both in terms of catalytic activity and of properties
of the polyene produced—high stereoregularity, high
molecular weight—and some of the reactions proceed
in a living fashion.
We have recently been interested in investigating the
catalytic properties of iridium organometallic deriva-
tives in alkynes polymerization. Few reports are found
in the literature on iridium catalysts which promote
carbon–carbon bond formation of alkynes, and only
oligomeric products are formed in the correspond-
ing reactions [21–24]. The water soluble compounds
Transition-metal catalyzed polymerization of
terminal alkynes (see Scheme 1) has recently at-
tracted considerable attention, due to the unique
physico-chemical properties of the polymeric
materials thus obtained. Photoconductivity, photolu-
minescence, oxygen permeability, humidity sensor,
ferromagnetism and non-linear optical properties
[1–4] are some of the properties of these polyenes,
which make them potentially important materials for
possible industrial applications. Accordingly, a variety
of catalytic systems have been devised for the poly-
merization of monosubstituted acetylenes. Molybde-
∗
Corresponding author. Tel.: +39-40-5583938;
fax: +39-40-5583903.
E-mail address: farnetti@dsch.univ.trieste.it (E. Farnetti).
1381-1169/$ – see front matter © 2003 Elsevier B.V. All rights reserved.
doi:10.1016/S1381-1169(03)00415-1

320
M. Marigo et al. / Journal of Molecular Catalysis A: Chemical 206 (2003) 319–329
2. Results and discussion
2.1. Catalytic reactions promoted by
HIr(cod)(PPh3)2
Scheme 1.
The first experiments were aimed at testing the
catalytic properties of HIr(cod)(PPh3)2 towards
C–C bond formation, using phenylacetylene (PA) as
monomer. By treating a boiling MeOH solution of the
iridium derivative with the alkyne ([PA]/[Ir] = 100),
formation of brown-red PPA was easily detected after
few minutes, due to the dark red color of the reaction
mixture, accompanied by precipitation of the higher
molecular weight fraction of the polyene. Analysis of
the reaction mixture by gas-chromatography revealed
that the monomer consumption was slowing down
with time and that after 5 h the reaction had stopped
(see Table 1 entry 5).
The polyene was isolated by filtration, and NMR
analysis of such material as well as of the final solution
revealed that trans-PPA was the only product formed.
Similar results were obtained in other solvents, such
as THF, CHCl3 and toluene, with the difference that,
together with the polyene, formation of small amounts
of the dimerization products (E)- and (Z)-butynenes
was observed (see table). When the catalytic reaction
was performed at lower temperature (40 ◦C, entries
1–3) the overall conversion lowered to 22–25%, and
trans-PPA was the only product formed. At variance,
an increase of the reaction temperature to 80 ◦C re-
sulted in a higher overall yield, but formation of the
IrCl(CO)(TPPTS)2 (TPPTS = P(m-C6H4SO3Na)3)
promote the formation of a low molecular weight
(Mn = 2 × 103) polyphenylacetylene (PPA) in low
yield [25], whereas in the reaction catalyzed by
IrH(CO)2(PPh3)2 the yield and Mn of the PPA formed
were not reported [26].
In our previous studies we employed as catalyst
precursors the dimeric compounds [Ir(cod)X]2 (cod:
1,5-cyclooctadiene; X: Cl, OMe), the rhodium ana-
logues of which are known to promote alkynes poly-
merization [27]. The organoiridium derivatives have
been found to catalyze phenylacetylene polymeriza-
tion at room temperature or lower, forming PPA with
either cis or trans stereochemistry, according to the
reaction temperature [28].
Here we report on the catalytic properties of
the compounds HIr(cod)(PR3)2 (PR3: PPh3, P(p-
MeOC6H4)3, P(o-MeOC6H4)Ph2, PCyPh2, PCy2Ph)
in the polymerization of phenylacetylene. The evolu-
tion of the catalyst precursor in the presence of the
monomer was followed by NMR spectroscopy, thus
providing useful information on the course of the
catalytic reaction.
Table 1
Polymerization of phenylacetylene catalyzed by HIr(cod)(PPh3)2
Entry
T (◦C)
Solvent
Conversion (%)
trans-PPAa
(E)-enynea
(Z)-enynea
1
2
3
4
5
6
7
8
9
40
40
40
60
60
60
60
60
80
80
110
THF
25
22
23
47
39
37
43
44
79
51
100
100
100
100
91
100
93
95
75
55
70
–
–
–
9
–
4
5
20
40
24
56
–
–
–
–
–
3
–
5
5
6
5
MeOH
CHCl3
THF
MeOH
EtOH
CHCl3
Toluene
Toluene
i-PrOH
Toluene
10
11
39
Experimental conditions: [Ir] = 3.4 × 10−3 mol l−1; [sub] = 0.34 mol l−1; [sub]/[Ir] = 100; Reaction time: 5 h.
a Product distribution (%).

M. Marigo et al. / Journal of Molecular Catalysis A: Chemical 206 (2003) 319–329
321
dimeric products was enhanced more than polymeriza-
tion; in refluxing toluene one of the oligomers, namely
the (E)-enyne, became the main reaction product.
Therefore, in order to obtain a higher conversion of
the polymeric product, a higher reaction temperature
was not appropriate. We therefore examined the effect
of the substituents of the phosphine ligands on the
catalytic reaction.
We also tried to prepare the hydrido bis-phosphino
compound with the ligand P(o-MeOC6H4)3: in this
case only small amounts of a monosubstituted com-
pound were formed, the 31P NMR spectrum (CDCl3)
of which consists in a singlet at δ + 10.50; in the
1H NMR spectrum signals corresponding to coor-
dinated cyclooctadiene at δ 4.69, 2.50 and 2.2–1.6
can be recognized, together with a singlet at 3.48
attributable to the o-methoxo group of the phosphine;
no signals at high field to TMS are detectable. Al-
though the presence of coordinated OMe group is not
apparent from NMR, the infrared spectrum shows a
strong signal at 1019 cm−1 which can be attributed
to Ir-OMe stretching; in fact, as the product is con-
taminated by free P(o-MeOC6H4)3, the resonance of
the methoxo group could be hidden by that of the
o-MeO group of the phosphine (3.75 ppm). There-
fore, the compound obtained can be formulated as
Ir(OMe)(cod)(P(o-MeOC6H4)3).
We also planned to prepare the hydrido-bisphosphino
compound with the ligand PCy3, however, with
such phosphine the reaction course was different.
A methanol suspension of [Ir(cod)(OMe)]2 and two
equivalents of PCy3 after 3 h under stirring pro-
duced a pale yellow solid, which was isolated. The
31P NMR spectrum of the compound consists of a
singlet at δ + 33.87, a chemical shift which sug-
gests formation of an Ir(III) species; high field 1H
NMR resonances (doublet of triplets at δ − 12.16,
JHP = 14.7 Hz and JHH = 4.4 Hz; triplet of triplets
at −12.82, JHP = 19.1 Hz, JHH = 4.4 Hz) together
with IR data (2060, 2035, 1925 and 1783 cm−1) in-
dicate formation of a hydrido-carbonyl compound.
The 1H NMR and IR data correspond to those re-
ported by Oro and co-workers [30] for the compound
2.2. Synthesis of the compounds HIr(cod)(L)2
The catalyst precursor HIr(cod)(PPh3)2 was syn-
thesized according to the procedure of Oro and
coworkers [29]: a suspension of the iridium dimer
[Ir(cod)(OMe)]2 in methanol was reacted at room
temperature with two equivalents of the phosphine,
yielding after 3 h the desired product as a white solid.
The same procedure was successfully employed
for the preparation of HIr(cod)(PR3)2, where PR3:
P(p-MeOC6H4)3, P(o-MeOC6H4)Ph2, PCyPh2,
PCy2Ph (reaction 1). Selected spectroscopic data for
the corresponding compounds are reported in Table 2.
12 [Ir(cod)(OMe)]2 + 2PR3
→ HIr(cod)(PR3)2 + CH2O
(1)
Attempts to prepare HIr(cod)(PR3)2 compounds with
other phosphine ligands were unsuccessful. The syn-
thesis of HIr(cod)(PEt2Ph)2 according to NMR data
of the crude product gave the desired compound (31P
NMR δ + 0.87 (s), 1H NMR δ − 14.07 (t, JPH
=
22.0 Hz)), which however we failed to isolate as a
crystalline material; purification by chromatographic
methods of the reaction mixture caused partial oxyda-
tion of the air-sensitive product.
Table 2
Selected NMR data for HIr(cod)(PR3)2.
PR3
31P
1H
P free
P coordinated
H-Ir
cod
PPh3
−5.50
−10.30
−16.90
−3.81
+7.31 (s)
+2.33 (s)
+3.89 (s)
+10.42 (s)
−0.43 (s)
−13.80(t, JHP = 22.0)
−13.82(t, JHP = 21.5)
−13.31(t, JHP = 22.5)
−14.00(t, JHP = 22.0)
−14.53(t, JHP = 23.5)
3.78; 3.46; 1.80; 1.52
3.73; 3.37; 1.78; 1.51
4.49; 4.04; 2.19; 2.01; 1.81
3.44; 3.11; 2.2–0.5
P(p-MeOC6H4)3
P(o-MeOC6H4)Ph2
PCyPh2
PCy2Pha
a
+2.97
4.61; 3.81; 2.6–0.9
Experimental conditions: CDCl3 solutions, 25 ◦C. Chemical shifts expressed in ppm; coupling constants expressed in Hz.
a C6D6 solution.

322
M. Marigo et al. / Journal of Molecular Catalysis A: Chemical 206 (2003) 319–329
mer-H3Ir(CO)(PCy3)2; these authors obtained the car-
bonyl species together with trans-H5Ir(PCy3)2; how-
ever, in our case the latter compound was not formed.
We were intrigued to find the hydrido-carbonyl com-
pound as the only reaction product, and repeated the
synthesis in a non-alcoholic solvent, in order to avoid
carbonylation: the reaction between [Ir(cod)(OMe)]2
and PCy3 was therefore repeated in CDCl3 in order
to follow its course by NMR. After 10 min at RT
only one product was present in solution as revealed
by 31P NMR, which showed a singlet at δ + 15.32
2.3. Catalytic reactions promoted by
HIr(cod)(PR3)2
The influence of the substituents of the phosphine
ligands on the catalytic properties of HIr(cod)(PR3)2
in promoting the polymerization of phenylacetylene
was examined using the compounds prepared with
the ligands P(p-MeOC6H4)3, P(o-MeOC6H4)Ph2,
PCyPh2, PCy2Ph. The results of the correspond-
ing catalytic reactions are summarized in Table 3.
When the ligand employed was an aromatic phos-
phine, no marked effect of the substituents on the
aromatic rings was observed on the catalytic ac-
tivity and product distribution, as well as on the
molecular weight of the polyene obtained. For this
series of precursors, the overall conversion was never
higher than 50%, and formation of the main prod-
uct trans-PPA was accompanied by small amounts
of enynes. An improvement in the catalytic activ-
ity was obtained by employing the adduct with the
ligand PCyPh2, with the conversion raising to up
to 68% and formation of the trans-polyene as the
main product. At variance, with the catalyst precursor
HIr(cod)(PCy2Ph)2 dimerization appeared to be the
main reaction, and trans-PPA was only a minor prod-
uct in the final mixture. In some of the catalytic re-
actions small amounts of the cyclotrimerization prod-
ucts, i.e. 1,3,5 and 1,2,4-triphenylbenzene, were also
detected.
With regard to the polyene formed, in all cases
stereoselective formation of trans-PPA was observed;
with the exception of the reactions catalyzed by
HIr(cod)(PCy2Ph)2, GPC molecular weight determi-
nation gave Mn values between 3000 and 5500, with
rather low polydispersion values (Mw/Mn = 1.4–1.6).
In all the catalytic reactions a deactivation process
was apparent, which was influenced to a large extent
neither by the solvent, nor by the phosphine coordi-
nated to iridium. Also when HIr(cod)(PCyPh2)2 was
employed as catalyst precursor, the reaction, although
initially faster than with other precursors, slowed down
markedly after few hours, i.e. in THF at 60 ◦C the
conversion, which after 5 h was 68%, did not exceed
75% after 24 h. With the same precursor, by changing
the [monomer]/[Ir] ratio from the usual value of 100
to 50, a higher conversion was obtained (77% in 5 h),
however the polyene obtained had a lower molecular
weight (Mn < 3000).
1
(PCy3 free, +11.01). The H NMR data correspond
to those reported [29] for Ir(OMe)(cod)(PCy3): at δ
4.36 and 2.79 the signals of the vinyl protons of cod,
at δ 2.2–1.1 the aliphatic resonances of coordinated
cod as well as those for PCy3, and at δ 3.46 a singlet
attributed to the methoxo group.
This product was slowly transformed with time
into another monophosphine derivative, which was
the only species in solution after 24 h. By addi-
tion of pentane it was possible to isolate this prod-
uct, which is probably the square planar compound
IrCl(cod)(PCy3). Such hypothesis is made on the ba-
sis of the following data: 31P NMR shows a singlet
1
at δ + 14.48; H NMR spectrum has resonances at-
tributable to coordinated cod and PCy3 (δ 4.78 and
3.15, vinyl protons of cod; δ 2.2–1.1, aliphatic protons
of cod and Cy), integrating for a PCy3:cod ratio of 1;
no hydridic signals are present, as well as resonances
which can be assigned to coordinated OMe group;
finally, the given formulation is in agreement with ele-
mental analysis data. The compound IrCl(cod)(PCy3)
is probably formed by substitution of the methoxo
group of Ir(OMe)(cod)(PCy3) with a chloride deriv-
ing from the solvent. It cannot be excluded that the
reaction proceeds via ␤-hydrogen elimination of the
methoxo group to give the hydride, with subsequent
fast substitution by a chloride: the latter reaction is
known to occur when iridium hydrides are dissolved
in chlorinated solvents [31].
By replacing CDCl3 with either benzene or THF,
from the reaction between [Ir(cod)(OMe)]2 and
PCy3 only Ir(OMe)(cod)(PCy3) was obtained: in
all the solvents employed it was never possible to
detect in solution the formation of the desired com-
pound HIr(cod)(PCy3)2, which is probably formed in
methanol, but is not observed due to the fast reaction
with the solvent to give mer-H3Ir(CO)(PCy3)2.

M. Marigo et al. / Journal of Molecular Catalysis A: Chemical 206 (2003) 319–329
323
Table 3
Polymerization of phenylacetylene catalyzed by HIr(cod)(PR3)2
Entry
PR3
Solvents
Conversion (%)
Dimersa
PPAa
Mn
Mw/Mn
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
PPh3
PPh3
PPh3
P(p-MeOC6H4)3
P(p-MeOC6H4)3
P(p-MeOC6H4)3
P(o-MeOC6H4)Ph2
P(o-MeOC6H4)Ph2
P(o-MeOC6H4)Ph2
PCyPh2
THF
47
39
43
44
38
39
39
45
35
68
66
55
48
18
21
9
–
5
8
5
10
15
5
91
100
95
90
95
90
75
95
80
90
90
90
30
20
20
3040
3860
4840
3220
5330
4330
3390
3440
3690
4120
4480
3660
2750
–
1.41
1.51
1.42
1.60
1.49
1.64
1.53
1.58
1.59
1.47
1.49
1.65
1.75
–
MeOH
CHCl3
THF
MeOH
CHCl3
THF
MeOH
CHCl3
THF
MeOH
CHCl3
THF
–
10
10
10
70
80
80
PCyPh2
PCyPh2
PCy2Ph
PCy2Ph
MeOH
CHCl3
PCy2Ph
–
–
Experimental conditions: [Ir] = 3.4 × 10−3 mol l−1; [sub] = 0.34 mol l−1; [sub]/[Ir] = 100; T = 60 ◦C. Reaction time: 5 h.
a Product distribution (%). Other products: 1,3,5 and 1,2,4-triphenylbenzene.
The effect of an addition of 1,5-cyclooctadiene to
the reaction mixture was also tested: when the reaction
with HIr(cod)(PPh3)2 was performed in THF at 60 ◦C
with excess cod ([cod]/[Ir] = 5) the conversion after
5 h was 51%, to be compared with 47% in the absence
of added diene, i.e. an almost negligible effect of added
diene to the catalytic reaction.
that an iridium–phosphine compound was also present
in the isolated solid. Such resonance is a doublet of
triplets at δ − 12.12, with JPH = 139.6 Hz typical of
a hydride trans to P, and JPH = 17.6 Hz attributable
to coupling of a hydride cis to two phosphorus atoms.
The 31P NMR spectrum revealed the presence of a
doublet at δ − 6.08 and a triplet at δ − 23.72 of rel-
ative intensity 2:1 (JPP = 13.9 Hz), which indicate
the presence of three phosphines coordinated in a
2.4. Isolation and characterization of
iridium-alkynyl complexes
fac-configuration. The aromatic resonances in the 13
C
NMR spectrum were almost completely obscured by
those of trans-PPA (130–126 ppm) present as a major
compound in the solid mixture, whereas two signals
were clearly detectable at higher field, i.e. a singlet at
The polyphenylacetylene produced in the catalytic
reactions was routinely isolated in order to determine
its stereochemistry and molecular weight. As the
polyene is insoluble in methanol, when the reactions
were performed in such solvent, the solid was recov-
ered by filtration. At variance, PPA was generally sol-
uble in other solvents employed (CHCl3, THF, C6H6),
therefore addition of an excess of methanol at the end
of the reaction caused precipitation of the polyene,
which was then recovered by filtration. In a typical
reaction performed with HIr(cod)(PPh3)2 as catalyst
precursor, the red-brown solid obtained was analyzed
by 1H and 13C NMR as well as GPC. In the 1H NMR
spectrum, aside the very broad resonance at δ 7.2 of
trans-PPA, the typical signals of the cis stereoisomer
[32] were never detected. Interestingly, in CDCl3 at
high field to TMS a multiplet was observed, indicating
δ 110.5 and a doublet of doublets at δ 95.9 (JCPtrans
=
24.8 Hz, JCPcis = 13.8 and 0 Hz) attributable to the
≡
C␤ and C␣ of Ir-C C–Ph, respectively. The presence
of the iridium-alkynyl mojety was confirmed by the
infrared spectrum, where aside the Ir-H stretching
at 2095 cm−1 two narrow bands appear at 2142 and
2119 cm−1, respectively. Therefore, the compound
isolated together with the polyene was identified as
≡
the iridium(III) derivative fac-HIr(C CPh)2(PPh3)3.
Such iridium(III) coordinatively saturated compound
is stable in CDCl3 solution under an inert atmosphere
for over 1 week, and does not behave as catalyst
precursor, as confirmed by its inactivity towards
phenylacetylene in experimental conditions similar

324
M. Marigo et al. / Journal of Molecular Catalysis A: Chemical 206 (2003) 319–329
1
to those employed in routine polymerization reac-
tions. Therefore, the iridium fraction which forms the
trisphosphino derivative does not participate to the
catalytic reaction.
reaction, which was monitored with time by H and
31P NMR, appeared to be slow at 25 ◦C: after 30 min
the major component of the mixture, according to 31
P
NMR, was still the starting compound (1), and only
small amounts of two new compounds were detected,
appearing as a broad signal at δ − 11 (2) and the set
Formation of the analogous compound HIr-
≡
(C CPh)2(P(p-MeOC6H4)3 was observed in the cat-
≡
alytic reactions employing the precursor with the
corresponding phosphine: this compound had 31P
NMR signals in CDCl3 consisting in a doublet at
δ − 10.93 and a triplet at −27.54 (JPP = 14.8 Hz),
whereas in the 1H NMR spectrum the hydridic signal
appeared as a doublet of triplets at δ − 12.29, with
JHPtrans = 140.9 and JHPcis = 17.6 Hz.
of signals corresponding to HIr(C CPh)2(PPh3)3 (3).
1
In the H NMR spectrum the hydridic signals of 1
and 3 were present at high field to TMS, whereas in
the low field region broad signals appeared between
3.9 and 3.3 ppm, which partially overlapped with the
resonances of cod vinyl protons of 1; moreover, small
signals of free cyclooctadiene were visible. By follow-
ing the reaction with time an increase in the signals of
2 and 3 was noticed in the 31P NMR spectrum, with
parallel increase of intensity of the broad signals in the
vinyl region of the 1H NMR spectrum. After 3 h from
the monomer addition, a series of spectra recorded
at lower temperatures revealed that the phosphorus
broad signal became a narrow singlet (δ − 11.47) at
At variance, a similar species was not present in
the isolated solid when the catalyst precursor was
HIr(cod)(PCyPh2)2: in this case neither 31P signals nor
1
high field resonances in the H NMR spectrum were
detected. However, we believe that the bis-alkynyl
compound is formed also in this case, but its lower
concentration (suggested by higher catalytic activity)
and/or higher solubility are the possible reasons why
it is not detected in the methanol insoluble fraction.
Finally, the product isolated at the end of the re-
actions promoted by HIr(cod)(PCy2Ph)2 contained an
iridium–phosphine species which showed only one
resonance in the 31P NMR spectrum (δ − 5.40 (s)):
this compound had no coordinated hydride accord-
−20 ◦C; accordingly, in the H NMR spectrum the
1
broad signals in the vinyl region resolved into two sig-
nals similar to those of coordinated cod of 1: such res-
onances at −20 ◦C have chemical shifts of 3.64 and
3.28 ppm. Moreover, even at low temperature no sig-
nals are present at high field of TMS, besides those of 1
and 3, indicating the absence of a hydride coordinated
to iridium for species 2. Unfortunately, the stability of
2 in solution did not allow to collect a 13C NMR spec-
trum, as decomposition occurred during the 12 h nec-
essary to obtain a reasonable signal/noise ratio. Com-
pound 2 cannot be unequivocally identified on the ba-
sis of the available spectral data, however we can rea-
sonably propose that compound 2 is the Ir(I) tetraco-
1
ing to H NMR and IR spectra, however in the lat-
ter a band at 2156 cm−1 suggests the presence of an
≡
Ir–C C–Ph moiety. The different catalytic behaviour
of HIr(cod)(PCy2Ph)2 in comparison to the other pre-
cursors might very well be related to a different chem-
istry of the iridium species in solution, probably due
to the increased bulk of the phosphine ligand.
=
ordinated species Ir(cod)(CH CHPh)(PPh3), formed
2.5. Spectroscopic studies
via insertion of the alkyne into the iridium-hydride
bond. The dependence of the NMR spectra on temper-
ature is probably due to exchange of the coordinated
phosphine with traces of the free ligand, which is cer-
tainly present in solution. The proton signals of the
A series of spectroscopic studies was performed on
the evolution of the catalyst precursors in the presence
of phenylacetylene, in order to shed light on the for-
mation of iridium derivatives during the catalytic re-
action, as well as on their role in the polymerization
catalysis. In the first experiment, a CDCl3 solution of
HIr(cod)(PPh3)2 (1) was treated with two equivalents
of phenylacetylene, by addition of the suitable amount
of a solution of the alkyne in CDCl3. After the addi-
tion the pale yellow solution turned bright yellow and
then orange, and after 30 min it had become red. The
=
CH CHPh [24,33] are most likely covered by the res-
onances of the aromatic protons of the three iridium
compounds which are present in the reaction mixture.
In a second experiment, a higher excess of the
monomer was employed ([PA]/[Ir] = 10) with the
aim to increase formation of 2 and 3, and to observe
the formation of the polyene. 31P NMR analysis of
the mixture revealed that formation of 2 and 3 at the

M. Marigo et al. / Journal of Molecular Catalysis A: Chemical 206 (2003) 319–329
325
Scheme 2.
expenses of 1 was nearly complete after 1 h; on the
other hand, the H NMR spectrum showed, together
with the signals of the three iridium compounds, the
broad resonance of trans-PPA.
oxidative addition of a second molecule of monomer
produces the iridium(III) derivative 3. The latter com-
pound is stable in deoxygenated solution for several
days, and it is catalytically inactive as demonstrated
in an experiment performed by employing isolated 3
as catalyst precursor.
1
From the results of the spectroscopic investigations
above reported, the evolution of the catalytic precursor
1 in the presence of the monomer is likely to follow
the course described in Scheme 2. Part of the iridium
centers undergoes dissociation of one phosphine with
formation of a square planar intermediate, which then
coordinates phenylacetylene in a ␲-fashion. Insertion
of the alkyne into the iridium-hydride bond produces
the vinyl compound 2, which is likely to be the ini-
tiator of the catalytic reaction. Another fraction of the
initial compound 1 picks up the phosphine released
in the former sequence as well as one molecule of the
monomer, with concomitant loss of cyclooctadiene:
the resulting ␲-alkyne intermediate eliminates H2,
presumably via alkyne oxydative addition. Finally,
2.6. Remarks on the reaction mechanism
Organorhodium compounds have been known for
the last 20 years to promote monosubstituted alkynes
polymerization with high catalytic activity, in all
cases with selective formation of the correspond-
ing cis-polyene. Investigations on the behavior of
rhodium-phosphine catalytic systems have demon-
strated that the presence of coordinated diolefin as
well as its nature are crucial factors for the polymer-
ization reaction: as a matter of fact, substitution of
the diene with two monoolefins leads to complete

326
M. Marigo et al. / Journal of Molecular Catalysis A: Chemical 206 (2003) 319–329
loss of polymerization catalysis. Such findings appear
to be also related to the catalytic properties of the
organoiridium precursors HIr(cod)(PR3)2, where the
presence of coordinated diene appears to be essential
for the catalytic reaction. With regard to iridium-based
catalysts, the dimeric compounds [Ir(cod)X]2 (X:
Cl, OMe) [28] show a similar behavior, with substi-
tution of cyclooctadiene with cyclooctene resulting
in a complete suppression of the polymerization
reaction.
When this manuscript was nearly completed we
became aware of a recent paper by Masuda and co-
workers [34], which also reports on iridium-catalyzed
polymerization of phenylacetylene. This group in-
vestigated on multi-component catalysts of the type
promote stereoselective polymerization to the trans-
polyene [5,36]. The interest towards trans-acetylenes
lies on the physico-chemical properties of such poly-
mers, particularly the non-linear optical properties,
which have been experimentally demonstrated to be
enhanced with increasing trans-conjugation length in
these polyenes [3].
In spite of the difference in the geometry of the
polyene obtained, it would be reasonable to be-
lieve that the polymerization reaction catalyzed by
HIr(cod)(PR3)2 proceeds via the insertion mecha-
nism proposed for rhodium-based catalysts [14]: the
stereochemistry observed might be the result of a
cis-insertion followed by an iridium-catalyzed isomer-
ization, or possibly of a trans-insertion as proposed
by Zhang and Yang for palladium-based catalysts
[37].
On the other hand, one cannot rule out the possibil-
ity that the catalytic reaction proceeds via a different
mechanism. Even if metathesis, which is operative for
molibdenum-promoted polymerization, would in prin-
ciple seem unlikely for the iridium catalysts, however
several iridium–carbenes are known, which are pos-
sible initiators of reactions which follow a metathesis
path; moreover, in the literature are reported [38–40]
some examples of iridacyclobutenes, similar to reac-
tion intermediates involved in the metathesis mecha-
nism.
=
[Ir(cod)Cl]2/norbornadiene/PPh3/Ph2C C(Ph)Li, and
found that the coordinated diene has a fundamental
role in the polymerization catalysis.
On the other hand, an important difference between
rhodium/phosphine and iridium/phosphine catalysts is
the stereochemistry of the polymerization reaction:
whereas rhodium promotes the selective formation of
cis-polyphenylacetylene, with the iridium-phosphine
compounds the trans-polyene is obtained.
In principle, thermal isomerization might be consid-
ered to be responsible for such difference, as the cat-
alytic reactions with iridium catalysts require a higher
temperature (60 ◦C) than those with rhodium (20 ◦C
or lower). Thermal cis-trans polyene isomerization is
known to occur at high reaction temperature, how-
ever the temperatures reported in the literature [1,35]
for such process are higher than those of our cat-
alytic reactions. We have tested the effect of a thermal
treatment of cis-polyene solutions in the same experi-
mental conditions used for the catalytic reactions, and
the result was that only a small amount (10–15%) of
isomerization occurred. Such findings indicate that the
polyene obtained in our catalytic reactions at 60 ◦C has
100% of trans geometry, as any amount of cis isomer
formed would not completely disappear by isomeriza-
tion, and therefore it should be detectable in the final
products.
3. Conclusions
The compounds HIr(cod)(PR3)2 (PR3: PPh3, P(p-
MeOC6H4)3, P(o-MeOC6H4)Ph2, PCyPh2) were
successfully employed as catalyst precursors for
the polymerization of phenylacetylene. The cat-
alytic reactions are highly selective, as only trans-
polyphenylacetylene is formed together with small
amounts of oligomerization products. The catalytic
activity suffers from a deactivation process, which
stops the reaction after a few hours. The evolution of
the catalyst precursor in the course of the catalytic
reaction has been investigated by spectroscopic stud-
ies, the results of which, together with the isolation
of catalytically inactive Ir(III) alkynyl derivatives, al-
lowed us to gain useful information both on the nature
of catalytically active species and on the deactivation
process.
Noteworthy, most of the catalytic systems which
promote alkynes polymerization either form the cis-
polyene (rhodium-based catalysts) or give rise to
polymers with variable cis-trans distribution along
the main chain (molibdenum-based catalysts); only
some examples of tungsten compounds are reported to

M. Marigo et al. / Journal of Molecular Catalysis A: Chemical 206 (2003) 319–329
327
4. Experimental section
ture was stirred at room temperature for 5 h, yielding a
white solid which was filtered, washed with methanol
and dried under vacuum. Yield 68%.
4.1. General
IR (Nujol): 2109 cm−1 (Ir-H). 31P {1H}NMR
(CDCl3, 25 ◦C): δ + 2.33 s.
All the reactions and manipulations were routinely
performed under an argon atmosphere using standard
Schlenk tube techniques.
1H NMR (CDCl3, 25 ◦C): δ 7.19 (m, 12H, m-Ar);
6.64 (d, 12H, o-Ar); 3.75 (s, 18H, OMe); 3.73 (bs, 2H,
=
=
C CH); 3.37 (bs, 2H, C CH); 1.78 (m, 4H, CH2); 1.51
Tetrahydrofuran (THF) was distilled over sodium
benzophenone ketyl just before use; benzene and
dichloromethane were distilled over CaH2, methanol
was distilled over CaO, and they were stored under
an inert atmosphere. Naphtalene was purified by re-
crystallization from ethanol. All the other chemicals
were reagent grade and were used as received by
commercial suppliers.
(m, 4H, CH2); −13.82 (t, 1H, Ir-H, JHP = 21.5 Hz).
4.4. Preparation of HIr(cod)(P(o-MeOC6H4)Ph2)2
A suspension of 80 mg (0.12 mmol) of [Ir(cod)-
(OMe)]2 in 26 ml of methanol was treated with
140 mg (0.48 mmol) of P(o-MeOC6H4)Ph2. The re-
sulting mixture was stirred at room temperature for
6 h, yielding a white solid which was filtered, washed
with methanol and dried under vacuum. Yield 48%.
IR (Nujol): 2134 cm−1 (Ir-H). 31P {1H}NMR
(C6D6, 25 ◦C): δ + 3.89 s.
The compounds [Ir(cod)(OMe)]2 [28] and HIr(cod)-
(PPh3)2 [29] were prepared according to the proce-
dures reported in the literature.
4.2. Instrumental
1H NMR (C6D6, 25 ◦C): δ 7.8–6.4 (m, 28H, Ar);
=
=
1H, 13C and 31P NMR spectra were recorded on
a JEOL EX400 spectrometer operating at 399.77,
100.54 and 161.82 MHz, respectively. 1H chemical
shifts are reported relative either to tetramethylsilane
(CDCl3 solutions) or to solvent peak (C6D6 solu-
tions); 13C chemical shifts are reported relative to
4.49 (bs, 2H, C CH); 4.04 (bs, 2H, C CH); 2.95 (s,
6H, OMe); 2.19 (bm, 2H, CH); 2.01 (bm, 2H, CH2);
1.81 (bm, 4H, CH2); −13.31 (t, 1H, Ir-H, JHP
=
22.5 Hz).
4.5. Preparation of HIr(cod)(PCyPh2)2
solvent peak (δ 77.0 for CDCl3, 128.0 for C6D6); 31
P
chemical shifts are reported relative to external 85%
H3PO4, with downfield shift positive. Infrared spec-
tra were recorded in Nujol mull on a Perkin-Elmer
System 2000 FT-IR spectrometer.
Chemical yields of the catalytic reactions were de-
termined by GLC on a Carlo Erba 6000 VEGA Series
2 equipped with a SE30 column, using naphtalene as
internal standard.
Molecular weight distributions of the polymers
were determined by GPC in CHCl3 at 25 ◦C on a
Milton Roy CM4000 instrument using a UV spec-
trometer detector operating at 270 nm, equipped with
CHROMPACK Microgel-5 columns.
A suspension of 100 mg (0.15 mmol) of [Ir(cod)-
(OMe)]2 in 32 ml of methanol was treated with 161 mg
(0.60 mmol) of PCyPh2. The resulting mixture was
stirred at room temperature for 5 h, yielding a white
solid which was filtered, washed with methanol and
dried under vacuum. Yield 75%.
IR (Nujol): 2127 cm−1 (Ir-H). 31P {1H}NMR
(CDCl3, 25 ◦C): δ + 10.42 s.
1H NMR (CDCl3, 25 ◦C): δ 7.3–6.8 (m, 20H, Ar);
=
=
3.44 (bs, 2H, C CH); 3.11 (bs, 2H, C CH); 2.2–0.5
(multiplets, 30H, CH2 (cod) and Cy); −14.00 (t, 1H,
Ir-H, JHP = 22.0 Hz).
4.6. Preparation of HIr(cod)(PCy2Ph)2
4.3. Preparation of HIr(cod)(P(p-MeOC6H4)3)2
A suspension of 66 mg (0.10 mmol) of [Ir(cod)-
(OMe)]2 in 15 ml of methanol was treated with 110 mg
(0.40 mmol) of PCy2Ph. The resulting mixture was
stirred at room temperature for 3 h, yielding a pale
A suspension of 80 mg (0.12 mmol) of [Ir(cod)-
(OMe)]2 in 24 ml of methanol was treated with 170 mg
(0.48 mmol) of P(p-MeOC6H4)3. The resulting mix-

328
M. Marigo et al. / Journal of Molecular Catalysis A: Chemical 206 (2003) 319–329
=
yellow solid which was filtered, washed with methanol
and dried under vacuum. Yield 48%.
(m, 2H, o-H(C6H5)), 5.84 (s, 1H, C CH); 13C {1H}
NMR (CDCl3) δ 142.9 and 139.3 (quaternary car-
IR (Nujol): 2142 cm−1 (Ir-H). 31P {1H}NMR
(C6D6, 25 ◦C): δ − 0.43 s.
bons), 131.8 (C CH), 127.8 and 127.5 (o- and m-Ar),
=
126.7 (p-Ar). For trans-polyphenylacetylene: 1H
NMR (CDCl3) δ 7.2 (very broad); 13C {1H} NMR
(CDCl3) δ 128 (very broad).
1H NMR (C6D6, 25 ◦C): δ 7.6–7.0 (m, 10H, Ar);
=
=
4.61 (bm, 2H, C CH); 3.81 (bm, 2H, C CH); 2.6–0.9
(multiplets, 52H, CH2 (cod) and Cy); −14.53 (t, 1H,
Ir-H, JHP = 23.5 Hz).
The yields of (E)-1,4-diphenyl-1-butyn-3-ene and
(Z)-1,4-diphenyl-1-butyn-3-ene were determined by
1H NMR (CDCl3) by integration of the signals of
the corresponding vinyl protons. For the (E) isomer:
δ 7.09 (d) and 6.43 (d) (JHH = 16.3 Hz); for the (Z)
isomer: δ 6.75 (d) and 5.97 (d) (JHH = 11.9 Hz).
Formation of 1,3,5-triphenylbenzene and 1,2,4-tri-
phenylbenzene was detected by 1H NMR, whereas
the relative amount of the two isomers was deter-
mined by integration of the 13C NMR signals of the
quaternary carbon atoms. For 1,3,5-triphenylbenzene:
1H NMR (CDCl3) δ 7.74, 7.46, 7.37; 13C NMR
(CDCl3) δ 142.3 and 141.1 (quaternary carbons),
4.7. Preparation of mer-H3Ir(CO)(PCy3)2
A suspension of 100 mg (0.15 mmol) of [Ir(cod)-
(OMe)]2 in 32 ml of methanol was treated with
168 mg (0.60 mmol) of PCy3. The resulting mixture
was stirred at room temperature for 3 h, yielding a
pale yellow solid which was filtered, washed with
methanol and dried under vacuum. Yield 79%.
IR (Nujol): 2060, 2035, 1925 cm−1 (Ir-H);
1783 cm−1 (CO).
1
129–125 (CH). For 1,2,4-triphenylbenzene: H NMR
31P {1H}NMR (CDCl3, 25 ◦C): δ + 33.87 s. 1H
NMR (CDCl3, 25 ◦C): δ 2.4–1.1 (multiplets, 66H, Cy);
−12.16 (dt, 2H, Ir-H, JHP = 14.7 Hz, JHH = 4.4 Hz);
−12.82 (tt, 1H, Ir-H, JHP = 19.1 Hz, JHH = 4.4 Hz).
(CDCl3) δ 7.84, 7.77, 7.46, 7.37; 13C NMR (CDCl3)
δ 141.4, 141.0, 140.9, 140.5, 140.3, 139.5 (quaternary
carbons), 129–125 (CH).
4.10. Determination of polyene molecular weight
4.8. Catalytic reactions in Schlenk tube
Determination of molecular weights via GPC was
performed on freshly prepared chloroform solu-
tions of the polymer. The number average molecular
weight (Mn) and polydispersion index (Mw/Mn) of the
polymers were calculated on calibrations using the
following polystyrene standards: Mp = 21,000 (Poly-
mer Laboratories), Mp = 9200 (Polysciences), Mp =
4000 (Aldrich), Mp = 980 (Polymer Laboratories).
A typical procedure is described as follows: a solu-
tion of HIr(cod)(PR3)2 (0.017 mmol) and of the GLC
standard naphtalene (100 mg) in 5.0 ml of THF (or
other solvent of choice) was heated to the desired reac-
tion temperature under inert atmosphere. Then 173 mg
of phenylacetylene (1.7 mmol, [sub]/[Ir] = 100) were
added. Samples were withdrawn from the reaction
mixture at time intervals, and disappearance of the
monomer was followed with time by GLC. The fi-
nal reaction mixture was treated with an excess of
methanol to precipitate the polymeric products; the
resulting red-brown solid was filtered, washed repeat-
edly with methanol and dried in vacuo.
Acknowledgements
The authors thank the Italian Ministero dell’ Uni-
versità e della Ricerca Scientifica e Tecnologica
(MURST) for finantial support. The authors are also
grateful to Johnson Matthey PLC for a loan of iridium.
4.9. Determination of product distribution and
stereochemistry
The stereochemistry of the polyphenylacetylene
References
1
obtained was determined by H and 13C NMR. For
1
cis-transoid polyphenylacetylene: H NMR (CDCl3)
[1] R. D’Amato, T. Sone, M. Tabata, Y. Sadahiro, M.V. Russo,
A. Furlani, Macromolecules 31 (1998) 8660.
δ 6.95–6.93 (m, 3H, m- and p-H(C6H5)), 6.64–6.62

M. Marigo et al. / Journal of Molecular Catalysis A: Chemical 206 (2003) 319–329
329
[2] Yu.P. Yampolskii, A.P. Korikov, V.P. Shantarovich, K.
Nagai, B.D. Freeman, T. Masuda, M. Teraguchi, G. Kwak,
Macromolecules 34 (2001) 1788.
[20] P. Mastrorilli, C.F. Nobile, A. Rizzuti, G.P. Suranna, D.
Acierno, E. Amendola, J. Mol. Catal. A: Chem. 178 (2002)
35.
[3] J.L. Bredas, C. Adant, P. Tackx, A. Persoons, Chem. Rev. 94
(1994) 243.
[21] C.S. Chin, G. Won, J. Song, Bull. Korean Chem. Soc. 15
(1994) 961.
[4] B.R. Weinberger, E. Ehrenfreund, A. Pron, A.J. Heeger, A.G.
MacDiarmid, J. Chem. Phys. 72 (1980) 4749.
[5] J. Kunzler, V. Percec, J. Polym. Sci. Part A: Polym. Chem.
28 (1990) 1221.
[22] S.-I. Lee, S.-C. Shim, T.-J. Kim, J. Polym. Sci. Part A: Polym.
Chem. 34 (1996) 2377.
[23] T. Ohmura, S. Yorozuya, Y. Yamamoto, N. Miyaura,
Organometallics 19 (2000) 365.
[6] S. Hayano, T. Masuda, Macromolecules 31 (1998) 3170.
[7] T. Masuda, S. Hayano, E. Iwawaki, R. Nomura, J. Mol. Catal.
A: Chem. 133 (1998) 213.
[8] T.A. Vijayaraj, G. Sundararajan, Organometallics 16 (1997)
4940;
G. Bhukta, R. Ramanujachary, G. Sundararajan, J. Organomet.
Chem. 601 (2000) 16.
[9] M.B. Mula, A.J. Beaumont, K.O. Doyle, M.L. Gallagher,
A.D. Rooney, J. Mol. Catal. A: Chem. 148 (1999) 23.
[10] R. Wang, F. Belanger-Gariepy, D. Zargarian, Organometallics
18 (1999) 5548.
[11] X. Zhan, M. Yang, J. Mol. Catal. A: Chem. 169 (2001) 27;
X. Zhan, M. Yang, H. Sun, J. Mol. Catal. A: Chem. 169
(2001) 63.
[12] A. Furlani, C. Napoletano, M.V. Russo, A. Camus, N.
Marsich, J. Polym. Sci. Part A 27 (1989) 75.
[13] M. Tabata, W. Yang, K. Yokota, J. Polym. Sci. Part A: Polym.
Chem. 32 (1994) 1113.
[24] C.S. Chin, H. Lee, H. Park, M. Kim, Organometallics 21
(2002) 3889.
[25] K.-S. Joo, S.Y. Kim, C.S. Chin, Bull. Korean Chem. Soc. 18
(1997) 1296.
[26] C.K. Brown, D. Georgiou, G. Wilkinson, J. Chem. Soc. (A)
(1971) 3120.
[27] M. Tabata, T. Sone, Y. Sadahiro, Macromol. Chem. Phys.
200 (1999) 265.
[28] M. Marigo, N. Marsich, E. Farnetti, J. Mol. Catal. A: Chem.
187 (2002) 169.
[29] M.J. Fernandez, M.A. Esteruelas, M. Covarrubias, L.A. Oro,
J. Organomet. Chem. 316 (1986) 343.
[30] M.J. Fernandez, M.A. Esteruelas, M. Covarrubias, L.A. Oro,
M.C. Apreda, C. Foces-Foces, F.H. Cano, Organometallics 8
(1989) 1158.
[31] E. Farnetti, G. Nardin, M. Graziani, J. Chem. Soc., Chem.
Commun. (1989) 1264.
[32] A. Furlani, S. Licoccia, M.V. Russo, A. Camus, N. Marsich,
J. Polym. Sci. Part A: Polym. Chem. 24 (1986) 991.
[33] H.E. Selnau, J.S. Merola, Organometallics 12 (1993)
3800.
[34] K. Kanki, Y. Misumi, T. Masuda, J. Polym. Sci. Part A:
Polym. Chem. 40 (2002) 1075.
[14] Y. Kishimoto, P. Eckerle, T. Miyatake, T. Ikariya, R. Noyori,
J. Am. Chem. Soc. 116 (1994) 12131;
Y. Kishimoto, P. Eckerle, T. Miyatake, M. Kainosho, A. Ono,
T. Ikariya, R. Noyori, J. Am. Chem. Soc. 121 (1999) 12035.
[15] Y. Misumi, T. Masuda, Macromolecules 31 (1998) 7572;
K. Kanki, Y. Misumi, T. Masuda, Macromolecules 32 (1999)
2384.
[35] F. Ishii, S. Matsunami, M. Shibata, T. Kakuchi, J. Polym.
Sci. Part B: Polym. Phys. 37 (1999) 16.
[16] Y. Goldberg, H. Alper, J. Chem. Soc., Chem. Commun. (1994)
1209.
[17] H. Katayama, K. Yamamura, Y. Miyaki, F. Ozawa,
Organometallics 16 (1997) 4497.
[36] T. Szymanska-Buzar, J. Mol. Catal. A 123 (1997) 113.
[37] X. Zhan, M. Yang, J. Mol. Catal. A: Chem. 169 (2001) 57.
[38] J.C. Calabrese, D.C. Roe, D.L. Thorn, T.H. Tulip,
Organometallics 3 (1984) 1223.
[18] J. Yao, W.T. Wong, G. Jia, J. Organomet. Chem. 598 (2000)
228.
[39] J.M. O’Connor, L. Pu, S. Woolard, R.K. Chadha, J. Am.
Chem. Soc. 112 (1990) 6731.
[19] M. Falcon, E. Farnetti, N. Marsich, J. Organomet. Chem. 629
(2001) 187.
[40] Y.-C. Cheng, Y.-K. Chen, T.-M. Huang, C.-I. Yu, G.-H. Lee,
Y. Wang, J.-T. Chen, Organometallics 17 (1998) 2953.